The for() loop ends up being in the code even if the log_debug() does
nothing. Add a condition to fix this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
This reverts commit 1fdf53ace1, reversing
changes made to e5aef1bbf1.
I had missed that this caused too much size growth on rcar3_salvator-x.
Signed-off-by: Tom Rini <trini@konsulko.com>
The for() loop ends up being in the code even if the log_debug() does
nothing. Add a condition to fix this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
When MMC booting fails it is sometimes hard to figure out what went
wrong as there is no error code. It isn't even clear which MMC device
was chosen, since SPL can have its own numbering.
Add some debugging to help with this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tim Harvey <tharvey@gateworks.com> says:
Modern eMMC v4+ devices have multiple hardware partitions per the JEDEC
specification described as:
Boot Area Partition 1
Boot Area Partition 2
RPMB Partition
General Purpose Partition 1
General Purpose Partition 2
General Purpose Partition 3
General Purpose Partition 4
User Data Area
These are referenced by fields in the PARTITION_CONFIG register
(Extended CSD Register 179) which is defined as:
bit 7: reserved
bit 6: BOOT_ACK
0x0: No boot acknowledge sent (default
0x1: Boot acknowledge sent during boot operation Bit
bit 5:3: BOOT_PARTITION_ENABLE
0x0: Device not boot enabled (default)
0x1: Boot Area partition 1 enabled for boot
0x2: Boot Area partition 2 enabled for boot
0x3-0x6: Reserved
0x7: User area enabled for boot
bit 2:0 PARTITION_ACCESS
0x0: No access to boot partition (default)
0x1: Boot Area partition 1
0x2: Boot Area partition 2
0x3: Replay Protected Memory Block (RPMB)
0x4: Access to General Purpose partition 1
0x5: Access to General Purpose partition 2
0x6: Access to General Purpose partition 3
0x7: Access to General Purpose partition 4
Note that setting PARTITION_ACCESS to 0x0 results in selecting the User
Data Area partition.
You can see above that the two fields BOOT_PARTITION_ENABLE and
PARTITION_ACCESS do not use the same enumerated values.
U-Boot uses a set of macros to access fields of the PARTITION_CONFIG
register:
EXT_CSD_BOOT_ACK_ENABLE (1 << 6)
EXT_CSD_BOOT_PARTITION_ENABLE (1 << 3)
EXT_CSD_PARTITION_ACCESS_ENABLE (1 << 0)
EXT_CSD_PARTITION_ACCESS_DISABLE (0 << 0)
EXT_CSD_BOOT_ACK(x) (x << 6)
EXT_CSD_BOOT_PART_NUM(x) (x << 3)
EXT_CSD_PARTITION_ACCESS(x) (x << 0)
EXT_CSD_EXTRACT_BOOT_ACK(x) (((x) >> 6) & 0x1)
EXT_CSD_EXTRACT_BOOT_PART(x) (((x) >> 3) & 0x7)
EXT_CSD_EXTRACT_PARTITION_ACCESS(x) ((x) & 0x7)
There are various places in U-Boot where the BOOT_PARTITION_ENABLE field
is accessed via EXT_CSD_EXTRACT_PARTITION_ACCESS and converted to a
hardware partition consistent with the definition of the
PARTITION_ACCESS field used by the various mmc_switch incarnations.
To add some sanity to the distinction between BOOT_PARTITION_ENABLE
(used to specify the active device on power-cycle) and PARTITION_ACCESS
(used to switch between hardware partitions) create two enumerated types
and use them wherever struct mmc * part_config is used or the above
macros are used.
Additionally provide arrays of the field names and allow those to be
used in the 'mmc partconf' command and in board support files.
The first patch adds enumerated types and makes use of them which
represents no compiled code change.
The 2nd patch adds the array of names and uses them in the 'mmc
partconf' command.
The 3rd patch uses the array of hardware partition names in a board
support file to show what emmc hardware partition U-Boot is being loaded
from.
Modern eMMC v4+ devices have multiple hardware partitions per the JEDEC
specification described as:
Boot Area Partition 1
Boot Area Partition 2
RPMB Partition
General Purpose Partition 1
General Purpose Partition 2
General Purpose Partition 3
General Purpose Partition 4
User Data Area
These are referenced by fields in the PARTITION_CONFIG register
(Extended CSD Register 179) which is defined as:
bit 7: reserved
bit 6: BOOT_ACK
0x0: No boot acknowledge sent (default
0x1: Boot acknowledge sent during boot operation Bit
bit 5:3: BOOT_PARTITION_ENABLE
0x0: Device not boot enabled (default)
0x1: Boot Area partition 1 enabled for boot
0x2: Boot Area partition 2 enabled for boot
0x3-0x6: Reserved
0x7: User area enabled for boot
bit 2:0 PARTITION_ACCESS
0x0: No access to boot partition (default)
0x1: Boot Area partition 1
0x2: Boot Area partition 2
0x3: Replay Protected Memory Block (RPMB)
0x4: Access to General Purpose partition 1
0x5: Access to General Purpose partition 2
0x6: Access to General Purpose partition 3
0x7: Access to General Purpose partition 4
Note that setting PARTITION_ACCESS to 0x0 results in selecting the User
Data Area partition.
You can see above that the two fields BOOT_PARTITION_ENABLE and
PARTITION_ACCESS do not use the same enumerated values.
U-Boot uses a set of macros to access fields of the PARTITION_CONFIG
register:
There are various places in U-Boot where the BOOT_PARTITION_ENABLE field
is accessed via EXT_CSD_EXTRACT_PARTITION_ACCESS and converted to a
hardware partition consistent with the definition of the
PARTITION_ACCESS field which is also the value used to specify the
hardware partition of the various mmc_switch incarnations.
To add some sanity to the distinction between BOOT_PARTITION_ENABLE
(used to specify the active device on power-cycle) and PARTITION_ACCESS
(used to switch between hardware partitions) create two enumerated types
and use them wherever struct mmc * part_config is used or the above
macros are used.
This represents no code changes.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Rather than having every caller set this up individually, create a
common init function. This allows new fields to be added without the
risk of them being left uninited.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Michael Trimarchi <michael@amarulasolutions.com>
At present spl_mmc_load() is the only caller of this function, passing
it a boot_device, an index into the available MMC devices. Pass the
device number instead, since it is known by the caller and simplifies
the code.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
Use 'ret' as the return code, since it may not be an error and this is
the common name in U-Boot. Make sure to return the error code when
given, rather than transforming it into -1 (-EPERM).
Signed-off-by: Simon Glass <sjg@chromium.org>
This check is not needed now, since printf() resolved to nothing if not
available. Drop the #ifdefs
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
As part of bringing the master branch back in to next, we need to allow
for all of these changes to exist here.
Reported-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
When bringing in the series 'arm: dts: am62-beagleplay: Fix Beagleplay
Ethernet"' I failed to notice that b4 noticed it was based on next and
so took that as the base commit and merged that part of next to master.
This reverts commit c8ffd1356d, reversing
changes made to 2ee6f3a5f7.
Reported-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
This follows the example of RISC-V where <asm/global_data.h> includes
<asm/u-boot.h> directly as "gd" includes a reference to bd_info already
and so the first must include the second anyhow. We then remove
<asm/u-boot.h> from all of the places which include references to "gd"
an so have <asm/global_data.h> already.
Signed-off-by: Tom Rini <trini@konsulko.com>
This converts the mmc loader to spl_load. Legacy images are handled by
spl_load (via spl_parse_image_header), so mmc_load_legacy can be
omitted. To accurately determine whether mmc_load_image_raw_sector is used
(which might not be the case if SYS_MMCSD_FS_BOOT is enabled), we introduce
a helper config SYS_MMCSD_RAW_MODE. This ensures we can inline spl_load
correctly when a board only boots from filesystems. We still need to check
for SPL_MMC, since some boards enable configure raw mode even without MMC
support.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Aligning addresses and sizes causes overhead which is unnecessary when we
are not loading from block devices. Remove bl_len when it is not needed.
For example, on iot2050 we save 144 bytes with this patch (once the rest of
this series is applied):
add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-144 (-144)
Function old new delta
spl_load_simple_fit 920 904 -16
load_simple_fit 496 444 -52
spl_spi_load_image 384 308 -76
Total: Before=87431, After=87287, chg -0.16%
We use panic() instead of BUILD_BUG_ON in spl_set_bl_len because we still
need to be able to compile it for things like mmc_load_image_raw_sector,
even if that function will not be used.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
For filesystems, filename serves the same purpose as priv. However,
spl_load_fit_image also uses it to determine whether to use a DMA-aligned
buffer. This is beneficial for FAT, which uses a bounce-buffer if the
destination is not DMA-aligned. However, this is unnecessary now that
filesystems set bl_len to ARCH_DMA_MINALIGN instead. With this done, we can
remove filename entirely.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Simplify things a bit for callers of spl_load_info->read by refactoring it
to use units of bytes instead of bl_len. This generally simplifies the
logic, as MMC is the only loader which actually works in sectors. It will
also allow further refactoring to remove the special-case handling of
filename. spl_load_legacy_img already works in units of bytes (oops) so it
doesn't need to be changed.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
dev and priv serve the same purpose, and are never set at the same time.
Remove dev and convert all users to priv. While we're at it, reorder bl_len
to be last for better alignment.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Remove NULL assignments to fields in spl_load_info when .load doesn't
reference these fields. This can result in more efficient code. filename
must stay even if it is unused, since load_simple_fit uses it.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
All "physical" addresses in SPL must be converted to virtual addresses
before access in order for sandbox to work. Add some calls to map_sysmem in
appropriate places. We do not generally call unmap_sysmem, since we need
the image memory to still be mapped when we jump to the image. This doesn't
matter at the moment since unmap_sysmem is a no-op.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Several SPL functions try to avoid performing initialization twice by
caching devices. This is fine for regular boot, but does not work with
UNIT_TEST, since all devices are torn down after each test. Add some
functions to invalidate the caches which can be called before testing these
load methods.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Make sure we have an IMX header before calling spl_load_imx_container,
since if we don't it will fail with -ENOENT. This allows us to fall back to
legacy/raw images if they are also enabled.
This is a functional change, one which likely should have been in place
from the start, but a functional change nonetheless. Previously, all
non-IMX8 images (except FITs without FIT_FULL) would be optimized out if
the only image load method enabled supported IMX8 images. With this change,
support for other image types now has an effect.
There are seven boards with SPL_LOAD_IMX_CONTAINER enabled: three with
SPL_BOOTROM_SUPPORT:
imx93_11x11_evk_ld imx93_11x11_evk imx8ulp_evk
and four with SPL_MMC:
deneb imx8qxp_mek giedi imx8qm_mek
All of these boards also have SPL_RAW_IMAGE_SUPPORT and
SPL_LEGACY_IMAGE_FORMAT enabled as well. However, none have FIT support
enabled. Of the six load methods affected by this patch, only SPL_MMC and
SPL_BOOTROM_SUPPORT are enabled with SPL_LOAD_IMX_CONTAINER.
spl_romapi_load_image_seekable does not support legacy or raw images, so
there is no growth. However, mmc_load_image_raw_sector does support loading
legacy/raw images. Since these images could not have been booted before, I
have disabled support for legacy/raw images on these four boards. This
reduces bloat from around 800 bytes to around 200.
There are no in-tree boards with SPL_LOAD_IMX_CONTAINER and AHAB_BOOT both
enabled, so we do not need to worry about potentially falling back to
legacy images in a secure boot scenario.
Future work could include merging imx_container.h with imx8image.h, since
they appear to define mostly the same structures.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
MMC devices do not have uclass platdata containing blk_descs, only their
child block devices do. Fortunately, we have a function just for this
purpose. This fixes subsequent calls to spl_mmc_load.
Fixes: bf28d9a659 ("spl: mmc: Use correct MMC device when loading image")
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Introduce two new weak functions, arch_spl_mmc_get_uboot_raw_sector() and
board_spl_mmc_get_uboot_raw_sector(), each of which can be overridden at
a matching level, that is arch/ and board/ , in addition to the existing
weak function spl_mmc_get_uboot_raw_sector().
This way, architecture code can define a default architecture specific
implementation of arch_spl_mmc_get_uboot_raw_sector(), while the board
code can override that using board_spl_mmc_get_uboot_raw_sector() which
takes precedence over the architecture code. In some sort of unlikely
special case where code has to take precedence over board code too, the
spl_mmc_get_uboot_raw_sector() is still left out to be a weak function,
but it should be unlikely that this is ever needed to be overridden.
Signed-off-by: Marek Vasut <marex@denx.de>
Rename this so that SPL is first, as per U-Boot convention. Also add
PAYLOAD_ since this is where in memory the parameters for the payload
have been stored.
Signed-off-by: Simon Glass <sjg@chromium.org>
When Falcon Mode is enabled, SPL needs to check the value of
CONFIG_SYS_MMCSD_RAW_MODE_ARGS_SECTOR. Unfortunately, it was using the
CONFIG_VAL(SYS_MMCSD_RAW_MODE_ARGS_SECTOR) which converts it into
CONFIG_SPL_SYS_MMCSD_RAW_MODE_ARGS_SECTOR when CONFIG_SPL_BUILD is
enabled. CONFIG_SPL_SYS_MMCSD_RAW_MODE_ARGS_SECTOR does not exist in
common/spl/Kconfig. Replaced with
defined(CONFIG_SYS_MMCSD_RAW_MODE_ARGS_SECTOR).
Signed-off-by: Elena Popa <elena.popa@nxp.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Mask macro PART_ACCESS_MASK filter out access bits of emmc register and
macro EXT_CSD_EXTRACT_BOOT_PART() extracts boot part bits of emmc register.
So use EXT_CSD_EXTRACT_BOOT_PART() when extracting boot partition.
Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
On some platforms is SYS_MMCSD_FS_BOOT_PARTITION unsupported. So allow to
completely disable MMC FS Boot support via new option SYS_MMCSD_FS_BOOT.
By default MMC FS Boot support is enabled (like it was before) except for
ARCH_MVEBU where MMC FS Boot supported is unsupported due to Marvell
BootROM limitations.
Signed-off-by: Pali Rohár <pali@kernel.org>
This is not needed and we should avoid typedefs. Use the struct instead
and rename it to indicate that it really is a legacy struct.
Signed-off-by: Simon Glass <sjg@chromium.org>
When attempting to load images from multiple MMC devices in sequence,
spl_mmc_load() chooses the wrong device from the second attempt onwards.
The reason is that MMC initialization is only done on its first call and
spl_mmc_load() will then continue using this same device for all future
calls.
Fix this by checking the devnum of the "cached" device struct against
the one which is requested. If they match, use the cached one but if
they do not match, initialize the new device.
This fixes specifying multiple MMC devices in the SPL's boot order to
fall back when U-Boot Proper is corrupted or missing on the first
attempted MMC device.
Fixes: e1eb6ada4e ("spl: Make image loader infrastructure more universal")
Signed-off-by: Harald Seiler <hws@denx.de>
This converts the following to Kconfig:
CONFIG_SYS_MMCSD_RAW_MODE_ARGS_SECTOR
CONFIG_SYS_MMCSD_RAW_MODE_ARGS_SECTORS
Signed-off-by: Tom Rini <trini@konsulko.com>
This was implemented in order to get dual-slot bootloader
partitions on the BeagleBone Black, whose MLO boots from
the first bootable partition: MLO chainloads u-boot in the
same way.
Signed-off-by: Jérôme Carretero <cJ-uboot@zougloub.eu>
Reviewed-by: Tom Rini <trini@konsulko.com>
Platforms can overwrite the weak definition of spl_mmc_boot_mode() to
determine where to load U-Boot proper from.
For most of them this is a trivial decision based on Kconfig variables,
but it might be desirable the probe the actual device to answer this
question.
Pass the pointer to the mmc struct to that function, so implementations
can make use of that.
Compile-tested for all users changed.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Stefano Babic <sbabic@denx.de>
Reviewed-by: Ley Foon Tan <ley.foon.tan@inte.com> (for SoCFPGA)
Acked-by: Lokesh Vutla <lokeshvutla@ti.com> (for OMAP and K3)
Reviewed-by: Simon Glass <sjg@chromium.org>
Falcon mode is very useful in improving boot speed. A question that
Falcon mode asks is "Where do I look for the kernel". With MMC boot
media, the correct answer is CONFIG_SYS_MMCSD_RAW_MODE_KERNEL_SECTOR.
The scope of this patch is to move this to Kconfig.
It is possible for a system to support Falcon mode from NOR but not
MMC. In that case, mmc_load_image_raw_os() would not be used. To
address this, conditionally compile mmc_load_image_raw_os() when
SPL_FALCON_BOOT_MMCSD, instead of SPL_OS_BOOT.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
[trini: Move spl_start_uboot to its own guard in spl_mmc.c, rerun migration]
Signed-off-by: Tom Rini <trini@konsulko.com>
Add parameter spl_boot_device to spl_parse_board_header(), which allows
the implementations to see from which device we are booting and do
boot-device-specific checks of the image header.
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Stefan Roese <sr@denx.de>
When building a system that has both TPL and SPL_OS_BOOT, code which
tests for CONFIG_SPL_OS_BOOT will be built and enabled in TPL, which is
not correct. While there is no CONFIG_TPL_OS_BOOT symbol at this time
(and likely will not ever be) we can use CONFIG_IS_ENABLED(OS_BOOT) in
these common paths to ensure we only compile these parts in the SPL
case.
Signed-off-by: Tom Rini <trini@konsulko.com>
Some image types (e.g. kwbimage v1) store the offset to SPL binary and
offset to U-Boot proper binary in their headers. To avoid reading SPL
binary when loading U-Boot proper, add support for specifying offset in
struct spl_image_info, which defines the offset from the beginning of
the header and the beginning of the executable data.
Initial support is added only for SPI, MMC and SATA code. We can extend
it later if needed.
Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Stefan Roese <sr@denx.de>
In general, Falcon mode means we're booting a linux kernel directly.
With FIT images, however, an OP-TEE secure kernel can be booted before
linux. Thus, if the next stage is an IH_OS_TEE, this isn't necessarily
a problem.
Of course, a general solution would involve mmc_load_image_raw_os()
only loading the binary, and leaving the decision of suitability to
someone else. However, a rework of the boot flow is beyond the scope
of this patch. Accept IH_OS_TEE as a valid OS value.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Factor out eMMC boot partition selection code into
default_spl_mmc_emmc_boot_partition() function and implement
weak spl_mmc_emmc_boot_partition(), so that architecture or
board code can override the eMMC boot partition selection.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Faiz Abbas <faiz_abbas@ti.com>
Cc: Harald Seiler <hws@denx.de>
Cc: Lokesh Vutla <lokeshvutla@ti.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Peng Fan <peng.fan@nxp.com>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Ye Li <ye.li@nxp.com>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
We should not be using typedefs and these make it harder to use
forward declarations (to reduce header file inclusions). Drop the typedef.
Signed-off-by: Simon Glass <sjg@chromium.org>
This function is only relevant to the MMC driver so calling it
spl_boot_partition() might be confusing. Rename it to
spl_mmc_boot_partition() to make its purpose more clear (and bring
it in line with spl_mmc_boot_mode()).
Signed-off-by: Harald Seiler <hws@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
The function's name is misleading as one might think it is used
generally to select the boot-mode when in reality it is only used by the
MMC driver to find out in what way it should try reading U-Boot Proper
from a device (either using a filesystem, a raw sector/partition, or an
eMMC boot partition).
Rename it to spl_mmc_boot_mode() to make it more obvious what this
function is about.
Link: https://lists.denx.de/pipermail/u-boot/2020-April/405979.html
Signed-off-by: Harald Seiler <hws@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
The call to spl_mmc_get_uboot_raw_sector() completely ignores and
overwrites the raw_sect value passed from the caller of spl_mmc_load().
Fix this by passing raw_sect to the function and returning the same
value in the default case.
Signed-off-by: Faiz Abbas <faiz_abbas@ti.com>
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
On Armada 38x platforms the ROM code loads SPL from offset 0 of eMMC
hardware boot partitions. When there are no boot partitions (i.e. SD
card) the ROM skips the first sector that usually contains the (logical)
partition table. Since the generated .kwb image contains the main U-Boot
image in a fixed location (0x140 sectors by default), we end up with the
main U-Boot image in offset of 1 sector. The current workaround is to
manually set CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR to 0x141 to
compensate for that.
This patch uses the run-time detected boot partition to determine the
right offset of the main U-Boot partition. The generated .kwb image is
now compatible with both eMMC boot partition, and SD card main data
partition.
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Reviewed-by: Stefan Roese <sr@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
